As coenzymes, NAD+ and NADH are important components of the respiratory chain, and involved in the electron transfer process in the respiratory chain (Rich, P. R. et al., Biochem Soc Trans. 2003, V.31 (6), pp. 1095-1105). In redox reactions of the respiratory chain, NAD+ acts as a proton carrier and transforms from its initial oxidation state to reduction state upon reception of an electron from other molecules; NADH, the product of this transformation, can act as a reducing agent providing electron for other molecules (Belenky. P. et al., Trends in Biochemical Sciences. 2007, V.32 (1), pp. 12-19). Recent studies have shown that, NAD(H) is not only involved in energy metabolism, substance synthesis, and antioxidation, but also relates to, inter alia, in vivo calcium homeostasis, gene expression, immunization, cell ageing and death, wherein NAD(H) plays vital roles. Accordingly, NAD(H) itself and numerous enzymes relating to NAD(H) metabolism have become targets for drug design (Sauve, A. A. et al., J Pharmacol Exp Ther. 2008, V.324(3), pp. 883-893).
However, in most living cells, the total amount of NAD(H) is about 10−6 M˜10−3 M, while the NAD+/NADH ratio also varies depending on intracellular states (Lin, S. J. et al., Current Opinion in Cell Biology. 2003, V.15(2), pp. 241-246), therefore, it is difficult to determine NAD(H). Earlier detecting methods mainly utilize the characteristic UV absorption of NADH at 340 nm, which leads to the UV spectrophotometry assay. This method has two main flaws: 1, the effective sensitivity is about 10−7 M, limited by the instrument precision; 2, effective differentiation between NADH and NADPH is not possible in complicated systems. A series of enzymatic assay are later developed based on the characteristic of NAD+ as a coenzyme which accepts an electron during electron transport and transforms to NADH. Other methods, such as HPLC analysis, the electrochemical assay, capillary electrophoresis, fluorescence imaging, etc., are also commonly reported in literatures. However, most of the methods either lack sensitivity towards target molecules in individual cells or lack capacity for localization to subcellular organelles. It is noteworthy that, a major common defect in these available methods is the need of sample processing including lysis, separation, and purification. As NADH itself is prone to oxidization, and errors are readily introduced, the cumbersome operations would lead to experimental results deviated from the bona fide values. In addition, these existing methods can not be applied to living animals or cells and can not detect in real time, which limits the applications in clinical diagnosis and prodrugs research. At present, NADH detection in living animals or cells can only be achieved by using NADH autofluorescence (Zhang, Q. H. et al., Science. 2002, V.295 (5561), pp. 1895-1897), but this traditional method has serious flaws as follows: first, it is known that the regulations of NAD+/NADH and NADP+/NADPH in cells are relatively independent, normally, NAD+/NADH ratio is about 700:1, while NADP+/NADPH ratio is about 1:200; second, the vast different in redox potential between NADH and NADPH indicates the distinct roles in energy metabolism and anabolism played by them; third, with NADH and NADPH autofluorescence being completely indistinguishable, the result obtained through autofluorescence imaging measurement is the sum of NADH and NADPH, and the data essentially indicates the concentration of protein-bound NADPH instead because the content of NADPH is low and mostly presented in protein binding form (Zhang, Q. H. et al., Science. 2002, V.295 (5561), pp. 1895-1897); fourth, since NADH is excited with wavelength in the ultraviolet range (340 nm) and its autofluorescence is weak, sophisticated and expensive equipments such as CritiView for clinical monitoring are required; furthermore, UV light has a rather weak capability of penetrating through tissues and can cause cell damages, so these optical properties severely restrict the application of autofluorescence monitoring.
Therefore, there is an urgent need in the art to develop a specific NADH detecting technique, especially, a specific technique which is suitable for detecting NADH in physiological level and subcellular level.
Relative to traditional detection techniques involving small molecule dye and rapid developing detection techniques using quantom dot, fluorescent protein detection technique has a unique overwhelming advantage in the imaging of most living cells; fluorescent protein can be genetically introduced into cells, tissues, and even whole organs, therefore it can be used as a whole-cell marker or gene activation indicator.
Green fluorescent protein is originally isolated from Aequorea victoria, and the wild-type AvGFP is consisted of 238 amino acids and has a molecular weight of about 26 kD. Recent study confirms that, in native GFP protein, three amino acids from 65 to 67, Ser-Tyr-Gly, are able to spontaneously form a fluorescent chromogenic moiety, wherein p-hydroxy-benzylidene-imidazolinone is the main luminous feature. The wild-type AvGFP has rather complex spectral characteristics with its main fluorescent excitation peak at 395 nm, and a secondary peak at 475 nm, whose amplitude intensity is approximately ⅓ of the main peak. Under standard solution condition, 395 nm excitation can produce 508 nm emission, and 475 nm excitation produces maximal emission at 503 nm (Heim, R. et al., Proc Natl Acad Sci USA. 1994, V.91 (26), pp. 12501-12504).
Upon intensive studies on GFP protein mutations, a variety of prominent GFP derivatives have been developed using molecular biotechnology. Through various single-point mutations or combination thereof made to the wild-type GFP, mutants such as enhanced-type GFP (S65T, F64L), YFP (T203Y) and CFP (Y66W) can be obtained. By rearranging GFP protein sequence to shift the original amino acids 145-238 to the N terminal and the amino acids 1-144 to the C terminal of the new protein, and binding the two fragments through a flexible short peptide chain, a space sensitive circular permutation fluorescent protein is formed thereby, and a T203Y point mutation thereupon results in a circular permutation yellow fluorescent protein cpYFP (Nagai, T. et al., Proc Natl Acad Sci U.S.A. 2001, V.98 (6), pp. 3197-3202).
Fluorescence-based analytical techniques have further developed along with the progression in fluorescent protein studies. One example is fluorescent resonance energy transfer (FRET) technique that is routinely adopted nowadays, the key mechanism of which is, when two fluorophores are in sufficiently close proximity, a donor entity absorbs photon of suitable frequency and is excited to a higher energy state returns to the ground state upon transferring energy to nearby acceptor entity via dipole-dipole interaction (that is, the occurrence of resonance energy transfer). FRET is a non-radiation energy transfer through intermolecular dipole-dipole interaction transferring energy from donor in excited state to result in acceptor in excited state, so that the fluorescence intensity of the donor decreases while the acceptor may emit characteristic fluorescence (sensitized fluorescence) which is stronger than its basic fluorescence, or it may emit no fluorescence (fluorescence quenching). Further studies of the green fluorescent protein show that cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP) derived from green fluorescent protein mutants constitute a prominent donor/acceptor pair. Emission spectrum of CFP substantially overlaps absorption spectrum of YFP, when CFP and YFP are in sufficiently close proximity and upon the excitation with the absorption wavelength of CFP, the chromophore of CFP will effectively resonance transfer energy to the chromophore of YFP, so CFP emitted fluorescence will be weakened or disappeared, and the main emission is YFP fluorescence. The efficiency of energy transfer between the two chromophores is inversely proportional to sixth power of the spatial distance between them, and is very sensitive to changes in spatial position. Therefore, existing studies report the use of genetically engineering recombinant methods for expression of a novel fusion protein having both termini of the protein of interest fused with CFP and YFP, respectively, such that spatial change caused by binding of the protein with its specific target molecule will be visualized by the fluorescent change.
So the fluorescent protein sequence used herein may come from Aequorea victoria fluorescent protein and its derivatives, including, but not limited to sequences of the following mutants: yellow fluorescent protein (YFP), green fluorescent protein (GFP), cyan fluorescent protein (CFP), and the likes, the sequence of yellow fluorescent protein YFP is preferable, and the sequence of circular permutation yellow fluorescent protein cpYFP is particularly preferable.
The technique described herein involves another protein, YdiH protein (also known as Rex protein), a bacterial transcriptional repressor protein already known in the art, which has a molecular weight of 23 kDa and can regulate fermentation and anaerobic respiration. Generally, YdiH proteins are derived from Thermus aquaticus (SEQ ID NO: 1 NCBI GenBank: AF061257.1), Streptomyces coelicolor (SEQ ID NO: 2 NCBI GenBank: AL9391.1) or Bacillus subtilis (SEQ ID NO: 3 NCBI GenBank: AL009126.3). YdiH protein is initially identified in 2003 from Streptomyces coelicolor by Brekasis and Paget et al., which is a redox-sensitive regulatory protein that widely presents in Gram-positive bacteria. The study on YdiH (Rex) protein of Streptomyces coelicolor indicates that it is a typical NAD(H) binding protein with Rossmann domain. The key Rossmann domain is a super-secondary protein structure mainly exists in nucleotide binding proteins, and it is a typical cofactor NAD(H) binding domain, represented by various cofactor NAD(H) binding proteins. The structure is essentially comprised of 6 β-pleated sheets linked through two pairs of α-helixes in the form of β-α-β-α-β. Since each Rossmann domain can bind one nucleotide molecule only, there are two Rossmann segments presented pairwise in dinucleotide binding proteins such as these for NAD. Current studies have shown that the Streptomyces coelicolor YdiH (Rex) protein can directly probe changes in cytoplasmic NADH/NAD+ ratio, while under aerobic conditions, the YdiH (Rex) protein can inhibit the transcription of its target genes (cydABC, nuoA-D and rexhemACD) when intracellular NADH/NAD+ ratio is at low level, but dissociates from its operon region at elevated NADH/NAD+ ratio, and during this dynamic process, the steric configuration of YdiH (Rex) protein transforms upon environmental change (Brekasis, D. et al., EMBO J., 2003, V.22 (18), pp. 4856-4865). Therefore, Rex protein is a good candidate for intracellular NADH sensor. Meanwhile, Wang et al. recently crystallized the Bacillus subtilis YdiH (Rex) protein and investigated its mechanism and function. Their results show that YdiH(Rex) protein from Bacillus subtilis is a homodimer protein having two functional domains, wherein the N-terminal domain (residues 1-85) is a DNA-binding domain, while the C-terminal domain (residues 86-215) is a typical Rossmann fold that can bind NADH (Wang, E. et al., Mol Microbiol 2008, V.69 (2), pp. 466-4′78).
Although YdiH (Rex) protein per se is sensitive to the redox state of the environment, the changes thereof are not intuitively exhibited and can not be captured externally. While by means of the fluorescent protein, we can ideally obtain a novel genetically encoded fluorescent sensor by fusioned expression of YdiH (Rex) and fluorescent protein, YdiH (Rex) is utilized for probing environmental redox state change and relaying the change to the fluorescent protein, which will visualize the change in environmental redox state in real-time and intuitively by the presence/absence or the intensity of the fluorescence generated thereby.
In summary, we believe that the use of recombinant fluorescent fusing protein which contains YdiH protein is able to meet the urgent need to detect NADH in physiological level and subcellular level.
The citation or discussion of any reference in this specification should not be construed as an admission that such reference is available as “Prior Art” to the present invention.